|
As this forum is rarely used anymore, we've locked it. Feel free to browse and read. Questions? Please reach out to us directly. Cheers! |
|
 |
|
|
| Author |
Message |
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sat Mar 08, 2008 3:59 pm Post subject: redo setup from scratch, something's not right - help? |
|
|
I am thoroughly befuddled right now. I reset everything and am starting a new setup from physical positioning on.
The problem I am having is that when I have the vertical centerline of the green grid centered on the screen, the distance of the grid from the left and right edges is not equal. There is "more" on the left than the right. The squares are bigger over there too.
Okay, I say, this is telling me that the PJ is shifted to the right of center and then rotated back to the left. I need to physically move it left, then rotate it right to square everything up.
Thing is, I have now gone left of center according to measurements taken by various means and I am still seeing the problem... so I am not sure what is going on.
When looking at green only, if the raster is centered on the tube face (it is), shouldn't everything be pretty close to symmetrical when the PJ is square to the screen?
I am starting to wonder if the green tube has been swung somehow, at some point...
Anyway, ideas would be greatly appreciated.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
 |
garyfritz
Joined: 08 Apr 2006
Posts: 12088
Location: Fort Collins, CO
|
| Posted: Sat Mar 08, 2008 4:05 pm Post subject: |
|
|
You're describing a problem with horizontal linearity. You adjust the linearity until the center-to-left-edge distance is the same as the center-to-right-edge distance. It's been a while since I've set up an 8500 so I don't remember where the control is, but it's probably under GEOM or PIC.
*DO NOT* correct this problem by moving the PJ!! You want the physical setup to be as perfect as possible: green CRT perfectly centered on the screen, PJ perfectly perpendicular to the screen, etc. Once you're certain the physical setup is perfect, THEN you tweak the raster centering, linearity, &etc on the screen. That will result in the least strain on the electronics and the most stable setup.
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sat Mar 08, 2008 4:12 pm Post subject: |
|
|
| garyfritz wrote: | You're describing a problem with horizontal linearity. You adjust the linearity until the center-to-left-edge distance is the same as the center-to-right-edge distance. It's been a while since I've set up an 8500 so I don't remember where the control is, but it's probably under GEOM or PIC.
*DO NOT* correct this problem by moving the PJ!! You want the physical setup to be as perfect as possible: green CRT perfectly centered on the screen, PJ perfectly perpendicular to the screen, etc. Once you're certain the physical setup is perfect, THEN you tweak the raster centering, linearity, &etc on the screen. That will result in the least strain on the electronics and the most stable setup. |
Ok. I'm familiar with the linearity adjustment, but I guess I thought I should be closer to even just on physical setup alone.
I guess it's more dragging the PJ around on the unistrut now to put it back. The upstairs neighbors hate that even more than when I watch action movies. 
Thanks.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sat Mar 08, 2008 4:33 pm Post subject: |
|
|
Okay, but wait...
If I have the projector centered and square to the screen, verified by measurement, shouldn't the green vertical grid centerline be in the center of my screen? 'Cause it isn't, it is right of screen center.
This is the problem I had from the time of my first setup, and I "corrected" it then by tweaking green convergence, but that doesn't seem right.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
Crabman
Joined: 17 Mar 2006
Posts: 197
Location: North West NC
|
| Posted: Sat Mar 08, 2008 5:54 pm Post subject: |
|
|
| lyd wrote: | Okay, but wait...
If I have the projector centered and square to the screen, verified by measurement, shouldn't the green vertical grid centerline be in the center of my screen? 'Cause it isn't, it is right of screen center.
This is the problem I had from the time of my first setup, and I "corrected" it then by tweaking green convergence, but that doesn't seem right.
lyd |
Did you adjust the raster center? What PJ are you using?
Clay
_________________
Hey Earl, there's a long skinny dog lying in the back seat. No wait, it's standing, it just doesn't have any legs.
Maybe it's a snake. Earl, do snakes have fur?
-Randy
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sat Mar 08, 2008 6:33 pm Post subject: |
|
|
Yeah, raster is centered, using the distance-to-edges method. Marquee 8500 is the projector.
I gave up on the all-measurement approach and and measured for center and then rotated the PJ to put the green vertical center on the screen center to get square. Went back and checked physical center, repeated until everything agrees. This unistrut mount is a real pain in the ass with the length of threaded rod I am using. Way too much play.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sat Mar 08, 2008 9:36 pm Post subject: |
|
|
ARGH!
Oh man, I don't know why I am having such problems here. The previous times I have done this went ok. Maybe I just didn't notice these things before...
Is it normal for the raster center to no longer be centered after you apply a source? I sure don't recall seeing this before. When I applied an external source at 1280x720p @ 96Hz, then flipped to the grid, the grid center is way to the left of screen center. Re-selected internal source #2 and it was back where it should be. Tried a couple of other resolutions/rates near where I want to be and it is shifted way left again.
I converged this thing a thousand times at 720p/96 over the last few months and sure never noticed the grid being way off center. What the heck?
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
AnalogRocks
Forum Moderator
Joined: 08 Mar 2006
Posts: 26706
Location: Toronto, Ontario, Canada
TV/Projector: Sony 1252Q, AMPRO 4000G
|
| Posted: Sat Mar 08, 2008 11:17 pm Post subject: |
|
|
| lyd wrote: | ARGH!
Oh man, I don't know why I am having such problems here. The previous times I have done this went ok. Maybe I just didn't notice these things before...
Is it normal for the raster center to no longer be centered after you apply a source? I sure don't recall seeing this before. When I applied an external source at 1280x720p @ 96Hz, then flipped to the grid, the grid center is way to the left of screen center. Re-selected internal source #2 and it was back where it should be. Tried a couple of other resolutions/rates near where I want to be and it is shifted way left again.
I converged this thing a thousand times at 720p/96 over the last few months and sure never noticed the grid being way off center. What the heck?
lyd |
I take it you're sync'ing your test patterns to your souce? Sounds like your sources video timings are off from the internal pattern generator's timings.
What source are you running into the projector?
_________________
Tech support for nothing
CRT.
HD done right!
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sat Mar 08, 2008 11:24 pm Post subject: |
|
|
| AnalogRocks wrote: | I take it you're sync'ing your test patterns to your souce? Sounds like your sources video timings are off from the internal pattern generator's timings.
What source are you running into the projector? |
Well... yeah, I mean the internal patterns have the source timings when you switch to them after first syncing to the source.
I'm using the same source as always, my htpc via hdmi into the moome card.
I just can't figure out what's going on. I guess I must have kludged something in my original setup, then forgot about it in subsequent tweaking, so I wasn't seeing this before. This is only the second time I have done a full reset, the first being when I got it.
Should I be re-centering the raster for the 720p timing, or is something deeply wrong here?
I wouldn't necessarily be so concerned, but when it shifts it is right off the phosphor on one side.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
AnalogRocks
Forum Moderator
Joined: 08 Mar 2006
Posts: 26706
Location: Toronto, Ontario, Canada
TV/Projector: Sony 1252Q, AMPRO 4000G
|
| Posted: Sat Mar 08, 2008 11:38 pm Post subject: |
|
|
Do you primaraly run 720P? If so then shift your raster's for that input. I'd rather have things not project off the sides of the tubes myself.
What other resolutions do you run from the HTPC?
_________________
Tech support for nothing
CRT.
HD done right!
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sun Mar 09, 2008 12:17 am Post subject: |
|
|
| AnalogRocks wrote: | Do you primaraly run 720P? If so then shift your raster's for that input. I'd rather have things not project off the sides of the tubes myself.
What other resolutions do you run from the HTPC? |
That's pretty much it at this point.
I'd be running 960p if I could get a custom res that would do it with this card/driver combo, but haven't had any luck there yet.
So, should I be planning to go back to the focus coils to make this raster shift, or just do it electronically? Oh... while we are on the subject, I have never figured out a way to shift the rasters electronically other than going into the guided setup and skipping through to that point. Is there a more direct way on the marquee?
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
AnalogRocks
Forum Moderator
Joined: 08 Mar 2006
Posts: 26706
Location: Toronto, Ontario, Canada
TV/Projector: Sony 1252Q, AMPRO 4000G
|
| Posted: Sun Mar 09, 2008 12:53 am Post subject: |
|
|
| lyd wrote: | | AnalogRocks wrote: | Do you primaraly run 720P? If so then shift your raster's for that input. I'd rather have things not project off the sides of the tubes myself.
What other resolutions do you run from the HTPC? |
That's pretty much it at this point.
I'd be running 960p if I could get a custom res that would do it with this card/driver combo, but haven't had any luck there yet.
So, should I be planning to go back to the focus coils to make this raster shift, or just do it electronically? Oh... while we are on the subject, I have never figured out a way to shift the rasters electronically other than going into the guided setup and skipping through to that point. Is there a more direct way on the marquee?
lyd |
I'm not a marquee guy (yet) I have AMPRO! lol
Is it the raster that's off the tube face or just the active image area?
_________________
Tech support for nothing
CRT.
HD done right!
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sun Mar 09, 2008 1:25 am Post subject: |
|
|
| AnalogRocks wrote: | | lyd wrote: | | AnalogRocks wrote: | Do you primaraly run 720P? If so then shift your raster's for that input. I'd rather have things not project off the sides of the tubes myself.
What other resolutions do you run from the HTPC? |
That's pretty much it at this point.
I'd be running 960p if I could get a custom res that would do it with this card/driver combo, but haven't had any luck there yet.
So, should I be planning to go back to the focus coils to make this raster shift, or just do it electronically? Oh... while we are on the subject, I have never figured out a way to shift the rasters electronically other than going into the guided setup and skipping through to that point. Is there a more direct way on the marquee?
lyd |
I'm not a marquee guy (yet) I have AMPRO! lol
Is it the raster that's off the tube face or just the active image area? |
The active image is barely touching one corner, raster is way off. I'm not sure what is going on right now, other than that things are way hosed up. It's been a long day, I'm going to watch a movie on my warped and only rough optical-focused display and try again tomorrow.
The one good thing that has come out of this is this convergence pattern I stumbled on. It kind of makes you go cross-eyed to use it, but my convergence has never been so tight. Blue especially is more dialed in than I have ever been able to get it before. I'll take what good news I can get right now.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
garyfritz
Joined: 08 Apr 2006
Posts: 12088
Location: Fort Collins, CO
|
| Posted: Sun Mar 09, 2008 4:51 am Post subject: |
|
|
| lyd wrote: | | Yeah, raster is centered, using the distance-to-edges method. |
If your linearity is off, that means your center crosshair is NOT in the "center" of your raster. Which means if you center the raster using the edges, the center crosshair is not in the center of the tube.
Do the mechanical setup perfectly. Get it **dead perpendicular** to the screen, NO rotation. Then the center of the **TUBE** will project directly in front of the green lens, which is presumably the center of your screen. If the center crosshair doesn't project to the center of your screen, that means the center crosshair isn't at the center of the tube, so you probably have a linearity problem. Re-center the raster so the center crosshair is perfectly centered on your screen. (Remember, this requires perfect mech setup first!!) Now with the center crosshair at the center of the screen, adjust your linearity so the left and right sides are the same size. Once you do that, your raster should be about the same distance-to-edges all the way around.
|
|
| Back to top |
|
|
Fujifrontier
Joined: 20 Oct 2007
Posts: 354
Location: San Antonio, Texas
|
| Posted: Sun Mar 09, 2008 6:07 am Post subject: |
|
|
|
that convergence pattern made me sick. how are you supposed to make sense of it?
|
|
| Back to top |
|
|
voodoo7869
Joined: 19 Oct 2007
Posts: 193
Location: Chicago, IL USA
|
| Posted: Sun Mar 09, 2008 6:43 am Post subject: |
|
|
did you check the green lens flapping?
_________________
Marquee 8000 modded, Marquee 9500lc with LUGs, Frankenyokes, and Moome. My avatar is not just a black box. CRT Is KING!!!!
|
|
| Back to top |
|
|
Elaine Benes
Joined: 25 Apr 2006
Posts: 1416
|
| Posted: Sun Mar 09, 2008 10:19 am Post subject: |
|
|
For a Marquee, you should always use the internal HELP guide and set it up EXACTLY how they tell you to set it up, having your source running the whole time while you do it.
There are some things about setting up a Marquee that are somewhat different than an NEC or a Sony or other projectors that use dynamic convergence primarily instead of zone convergence.
There's a sequence in the Marquee setup where it tells you to get your projector setup correctly you need to pivot your projector around an imaginary center point to achieve a symmetrical and centered image on your screen. This differs from *many* setup theories in that most recommend you have everything dead square/centered/perpendicular to the screen and then adjust using electronic controls. The Marquee doesn't require this to be the case. The exact phrase from my ancient 8000's help screen is " Pivot the projector and move it side to side AS REQUIRED to display a symmetrical crosshatch centered left to right on the screen". The caveat is added: " Note: Pivot around an imaginary vertical axis through the rear wing nut of the green lens to avoid changing the projector to screen distance." They're concern is primarily the correct throw distance for proper physical focus. As long as that is maintained, lens flapping and center/edge focus can focus your image regardless of your projectors exact orientation to the screen.
An NEC, for example, does not have continuously variable lens flapping, so NEC recommends you have your projector EXACTLY centered, squared, and perpendicular to your screen in order to be able to properly focus the image on the screen at all points.
Marquee setup isn't too concerned about absolute exact physical placement, but more with raster centering on the tubeface, then projector placement in relation to screen to produce a "symmetrical crosshatch centered left to right on the screen".
I believe this setup theory is pretty much particular to Marquee's, btw, but it works...
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sun Mar 09, 2008 12:59 pm Post subject: |
|
|
| garyfritz wrote: | | lyd wrote: | | Yeah, raster is centered, using the distance-to-edges method. |
If your linearity is off, that means your center crosshair is NOT in the "center" of your raster. Which means if you center the raster using the edges, the center crosshair is not in the center of the tube.
Do the mechanical setup perfectly. Get it **dead perpendicular** to the screen, NO rotation. Then the center of the **TUBE** will project directly in front of the green lens, which is presumably the center of your screen. If the center crosshair doesn't project to the center of your screen, that means the center crosshair isn't at the center of the tube, so you probably have a linearity problem. Re-center the raster so the center crosshair is perfectly centered on your screen. (Remember, this requires perfect mech setup first!!) Now with the center crosshair at the center of the screen, adjust your linearity so the left and right sides are the same size. Once you do that, your raster should be about the same distance-to-edges all the way around. |
Oh... hmm. Yeah, that makes sense, Gary.
I have a real problem with total perfection with the physical setup in my environment, however. Between the unistrut mount and a severe lack of parallel and square surfaces to measure from, positioning is extremely painful. It is difficult to have 100% confidence in dead perpendicularity. I don't think I can achieve more than +/- 1/2" on centerline and +/- a couple degrees of rotation right now.
A large part of the problem with that is a slightly non-linear horizontal concavity in my Wilsonart screen. It hasn't hurt focus, heck it might even help a bit, but it makes measuring square off of it tricky at best. My next semi-DIY screen will be Da-Lite Cinema Vision material on a frame of manufactured [URL=http://tinyurl.com/yu3sts
]museum-grade stretcher bars[/URL] (should run around $300 - $400 total for a tensioned 96" 16:9, seems like an OK value) and will solve at least that one problem, but that's a bit down the road.
In the short term, I guess I will have to continue to kludge around this issue.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sun Mar 09, 2008 1:04 pm Post subject: |
|
|
| Fujifrontier wrote: | | that convergence pattern made me sick. how are you supposed to make sense of it? |
Yeah, like I said, it does make you go a bit cross-eyed. It is easier to focus on, a bit anyway, when you drop to just R+G and B+G to do the convergence, though, and it sure helped me do a really tight job. Really helps you spot those small mis-convergences that it seems you have to start tweaking three zones over from the trouble spot to get solved. And like I said, blue especially has never been so good.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
lyd
Joined: 15 Sep 2007
Posts: 390
Location: Lake Mills, Wi
|
| Posted: Sun Mar 09, 2008 1:10 pm Post subject: |
|
|
| voodoo7869 wrote: | | did you check the green lens flapping? |
I don't think flapping would cause this problem, would it? That's just going to effect the edge focus, and I suppose geometry a bit. I do need to go back and do the scheimpflug on all three lenses now that I have shifted the projector, which is a job for this morning, but I don't expect it to change the non-linearity issue.
I did briefly consider the possibility that the green gun had been rotated off square to the projector frame, but I don't know how to check that, and anyway I think Gary's point about the centerline not being in the center of the raster is likely the correct answer.
lyd
_________________
de gustibus non disputandum
|
|
| Back to top |
|
|
|
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
|
Forum powered by phpBB © phpBB Group
|
|
